1. Chapter 8 Delving Into The Use of Inference 8.1 Estimating with Confidence 8.2 Use and Abuse of Tests

2. 2 Ch8.1 Eapter 6 Introduction to Inference8. How Confidence Intervals Behave Choosing a Sample Size Two-Sided Significance Tests and Confidence Intervals Some Cautions 8.1 Estimating with Confidence

3. After we have selected a sample, we know the responses of the individuals in the sample. However, the reason for taking the sample is to infer from that data some conclusion about the wider population represented by the sample. 3 Statistical Inference Statistical inference provides methods for drawing conclusions about a population from sample data. Population Sample Collect data from a representative sample... Make an inference about the population.

4. Confidence Level The confidence level is the overall capture rate if the method is used many times. The sample mean will vary from sample to sample, but when we use the method estimate ± margin of error to get an interval based on each sample, C% of these intervals capture the unknown population mean µ. To say that we are 95% confident is shorthand for “95% of all possible samples of a given size from this population will result in an interval that captures the unknown parameter.” Interpreting a Confidence Level 4

5. 5 Confidence Interval A level C confidence interval for a parameter has two parts:  An interval calculated from the data, which has the form: estimate ± margin of error  A confidence level C, which gives the probability that the interval will capture the true parameter value in repeated samples. That is, the confidence level is the success rate for the method. A level C confidence interval for a parameter has two parts:  An interval calculated from the data, which has the form: estimate ± margin of error  A confidence level C, which gives the probability that the interval will capture the true parameter value in repeated samples. That is, the confidence level is the success rate for the method. We usually choose a confidence level of 90% or higher because we want to be quite sure of our conclusions. The most common confidence level is 95%. estimate ± margin of error The Big Idea: The sampling distribution of tells us how close to µ the sample mean is likely to be. All confidence intervals we construct will have a form similar to this:

6. 6 How Confidence Intervals Behave The t confidence interval for the mean of a population illustrates several important properties that are shared by all confidence intervals in common use.  The user chooses the confidence level and the margin of error follows.  We would like high confidence and a small margin of error.  High confidence suggests our method almost always gives correct answers.  A small margin of error suggests we have pinned down the parameter precisely. The margin of error for the t confidence interval is: The margin of error gets smaller when:  t* gets smaller (the same as a lower confidence level C).  s is smaller. It is easier to pin down µ when s is smaller.  n gets larger. Since n is under the square root sign, we must take four times as many observations to cut the margin of error in half. The margin of error for the t confidence interval is: The margin of error gets smaller when:  t* gets smaller (the same as a lower confidence level C).  s is smaller. It is easier to pin down µ when s is smaller.  n gets larger. Since n is under the square root sign, we must take four times as many observations to cut the margin of error in half.

7. Impact of Sample Size Sample size n Standard error  ⁄ √n The spread in the sampling distribution of the mean is a function of the number of individuals per sample.  The larger the sample size, the smaller the standard deviation (spread) of the sample mean distribution.  The spread decreases at a rate equal to √n. 7

8. 8 Choosing the Sample Size (Proportion) In planning a study, we may want to choose a sample size that allows us to estimate a population proportion within a given margin of error. z* is the standard Normal critical value for the level of confidence we want.

9. 9 Sample Size for Desired Margin of Error for a Proportion p The level C confidence interval for a proportion p will have a margin of error approximately equal to a specified value m when the sample size satisfies Here, z * is the critical value for confidence level C, and p * is a guessed value for the proportion of successes in the future sample. The margin of error will be less than or equal to m if p * is chosen to be 0.5. Then the sample size is

10. 10 Suppose you wish to determine what percent of voters favor a particular candidate. Determine the sample size needed to estimate p within 0.03 with 95% confidence. The critical value for 95% confidence is z* = 1.96. Since the company president wants a margin of error of no more than 0.03, we need to solve the equation: We round up to 1068 respondents to ensure the margin of error is no more than 0.03 at 95% confidence. Example

11. Choosing the Sample Size (Mean) You may need a certain margin of error (e.g., drug trial, manufacturing specs). In many cases, the population variability (  is fixed and unknown, but we can choose the number of measurements (n) – sometimes we must estimate  a pilot study. The confidence interval for a population mean (sigma known) will have a specified margin of error MOE when the sample size is: Remember, though, that sample size is not always stretchable at will. There are typically costs and constraints associated with large samples. The best approach is to use the smallest sample size that can give you useful results. 11

12. Confidence intervals to test hypotheses Because a two-sided test is symmetrical, you can also use a confidence interval to test a two-sided hypothesis. If the hypothesized value of the mean is not inside the 100*(1-α) % confidence interval, then reject the null hypothesis at the α level, assuming a two-sided alternative. In a two-sided test, C = 1 – α, where C confidence level and α = significance level

13. Some Cautions  The data should be a SRS from the population.  The formula is not correct for other more complex sampling designs.  Inference cannot rescue badly produced data.  Confidence intervals are not resistant to outliers.  If n is small (<15) and the population is not Normal, the true confidence level will be different from C. (Recall the “guidelines” regarding the use of the t-statistic for different values of n)  The margin of error in a confidence interval covers only random sampling errors! 13

14. 8.2 Use and Abuse of Significance Tests  Choosing a Significance Level  What Statistical Significance Does Not Mean  Don’t Ignore Lack of Significance  Type I and Type II Errors  Error Probabilities 14

15. Choosing the Significance Level  Factors often considered:  What are the consequences of rejecting the null hypothesis?  Are you conducting a preliminary study? If so, you may want a larger  so that you will be less likely to miss an interesting result. Cautions about Significance Tests Some conventions:  We typically use the standards of our field of work.  There are no “sharp” cutoffs: for example, 4.9% versus 5.1%.  It is the order of magnitude of the P-value that matters: “somewhat significant,” “significant,” or “very significant.” 15

16. What Statistical Significance Does Not Mean Statistical significance only says whether the effect observed is likely to be due to chance alone because of random sampling. A statistically significant result may not be practically significant. That’s because statistical significance doesn’t tell you about the magnitude of the effect, only that there is one. An effect could be too small to be relevant. And with a large enough sample size, significance can be reached even for the tiniest effect.  A drug to lower temperature is found to reproducibly lower patient temperature by 0.4°Celsius (P-value < 0.01). But clinical benefits of temperature reduction only appear for a 1° decrease or larger. Cautions about Significance Tests 16

17. Don’t Ignore Lack of Significance  Consider this provocative title from the British Medical Journal: “Absence of evidence is not evidence of absence.”  Having no proof of who committed a murder does not imply that the murder was not committed. Indeed, failing to find statistical significance in results is not rejecting the null hypothesis. This is very different from actually accepting it. The sample size, for instance, could be too small to overcome large variability in the population. When comparing two populations, lack of significance does not imply that the two samples come from the same population. They could represent two very distinct populations with similar mathematical properties. Cautions about Significance Tests 17

18. Beware of Searching for Significance There is no consensus on how big an effect has to be in order to be considered meaningful. In some cases, effects that may appear to be trivial can be very important. Example: Improving the format of a computerized test reduces the average response time by about 2 seconds. Although this effect is small, it is important because this is done millions of times a year. The cumulative time savings of using the better format is gigantic. Always think about the context. Try to plot your results, and compare them with a baseline or results from similar studies. Cautions about Significance Tests 18

19. HW: Read sections 8.1 & 8.2 up through p.439 (Type I and Type II errors...) Work on #8.5, 8.6, 8.8-8.12, 8.16, 8.21, 8.32, 8.49, 8.51, 8.57 19